My CNC project

[eitanwaks]

Hi everybody,
after being on the sidelines for quite a while I've decided to take to the field. I have read it and learned a lot and have a lot more to learn but also I feel that I am ready to take the leap and start designing my own CNC router.

I am the proud owner of the following items:
rails:
3 x Star 1605-10 1416 mm long with 2 1651 type runner blocks per rail
1 x HSR35 515 mm long
1 x HSR35 470 mm long
2 x LWHG45 838 mm long

ball screws:
1 x Kuroda precision ballscrew d15 l10 760 mm with support blocks and motor mount
1 x Kuroda precision ballscrew d15 l10 560 mm with support blocks and motor mount
1 x Kuroda precision ballscrew d15 l10 280 mm with support blocks and motor mount

spindle:
Kavo 4052
Kavo 4444

motors:
3 x NEMA 34 640 oz. stepper

drivers:
3 x Gecko 201

Breakout board:
Campbell


after doing some research this is a design that I came up with. I would love to get any input and suggestions on how to make my design better.

Thank you for all of your help

[Piobaire]

Looks very good, but I think the plate that holds the spindle unit looks a bit weak, as the rails are not supporting the last bit of the spindle plate. I think you should try and make a box-design of the spindle plate with the spindle unit inside. Have you looked at a servo kit from Ajax instead of steppers ? www.ajaxcnc.com


BR
/Piobaire

[eitanwaks]

Piobaire,
the items listed in my original post have already been purchased therefore I think that buying a new set of servo motors for my project would be redundant. I have a look at the web site you mentioned (Ajax). It seems as they have a lot of professional equipment however, none that interest me currently.

I agree that the plate that holds up the spindle is weak. Could you please explain to me what a box design is? What other options do I have if I want to maintain the full travel in this direction?

[Piobaire]

I understand the trouble with getting new parts. I also have a project of making a home made retrofit and must think more then twice about expences.

A box design would you get if you added two plates beside the spindle motor and one wider plate on top of them. Then the spindle is inside a box which will reinforce the spindle area alot. As the force added on the spindle when the table moves (X-direction?) create a torque on the plate thickness. The premitted torque on the plate is dependent on the thickness in cubic. If you make a box design (don't need a major thickness of the material as long you could fit in scres and pins) it would be a great reinforcement.

The weak section is just above the spindle motor where the rails ends and the plate decreases in width. There are only the plate thickness (10mm?) holding the spindle forces.
BR
/T

[Piobaire]

Oh, sorry!
The rails are fitted in the other part so the spindle plate is weak all over, except where the waggons are. reinforce it with two bars on each side of the spindle motor and hold them together on the opposite side of the spindle plate just above the spindle motor attachment and upwards abit. Have you got the design i CAD (preferable SolidWorks, Pro/E or Catia) I could give you a hint if I got some spare time.

BR
/T

[Jim DuBois]

I guess a question of you at this time is what do you want to machine using your "CNC Router"? I ask because it appears as if your design is very robust in having a double legged "gantry" as well as "over engineered" in the rails and bearings as you have drawn your machine. Nice job on the drawings by the way.

Also, as designed your machine is not very efficent from a use of space perspective....you are consuming a bunch of square feet of benchtop for not very much travel.

Finally, I would be concerned as to chips and dust contamination of the rails and ballscrews.....back to the question of intended use? What are your required tolerances on the parts you plan to make? There is little need to make a machine that holds a tolerance of say .001" if you are machining wood.

[eitanwaks]

Piobaire,
I will gladly send you the file (solid works) or maybe uploaded here. It will be in zip format. I still don't quite understand the meaning of the box. After running Cosmos I realize that the plate can hold about 40 kg of force at the spindle tip. I don't know whether that is enough. I will have to calculate the forces on the drill bits in order to be sure.

Jim DuBois,
my motivation to build a CNC router stems from two places initially. First and foremost, due to my disability (quadriplegic (wheelchair-bound)) I'm in constant need of adapting equipment so that I can use it. Among the projects have done up until now have been tripods for my cameras, brakes for my wheelchair, various accessories for an adaptable bicycle and more. I'm sick and tired of having to go half an hour to an hour to shops around town and asking them for favors or commissioning them for work. The problem isn't getting shops rather waiting weeks on hand for them to finish the work or not even accepting to do the job because it isn't worth their time. The second reason which actually was a straw that broke the camel's back is that I'm building a speaker system for my home. As a perfectionist I studied acoustics and arrived at a speaker designed which would be ideal for my needs. The problem is it had to be manufactured. Several parts needed to be manufactured with a CNC machine. The tolerances on these parts are quite tight. After going to several places the average cost for manufacturing the parts came to about $1500. I was hoping that it would be a lot less. I eventually came to the conclusion that I could probably build a machine that would build as many parts as I want for not much more than that. From then forward I had been researching CNC design. This is almost a year now.

The fact that you called my design over engineered is a complement considering I'm a student of Mechanical Engineering in the Technion University in Israel. Thank you for your comments about the drawings. This project sure has honed my solid works skills. Like most of us (I assume) I purchased most of the parts used, thus my design had to be designed around those constraints.

I opted for a moving table design instead of the gantry because I presume that moving table would be easier on the Stepper motors that moving the gantry. I also believe that a moving table design is more stable.

Regarding chips and contamination, I will have to figure out some sort of bellows design to cover the rails. As I mentioned before, I will probably do many parts with quite close tolerances. In addition, if the machine turns out well I can submit it as a project for my studies. For the project I intend to add two more axis to make it a five axis design but first things first.

[Piobaire]

It's not the actual load that are of most concern, however 400N at the tool is not much, but I think it have to due in this case. I also think the rest of the machine in Al-profiles are a bit weak for steel cutting purposes. Problem that you'll get with too weak design is vibrrrrrrrrations and that also causes the tool to cut badly and increases tool wear.

Have a look at this nice and well priced machines, I think yours are about the same in general as the MDX-650.

http://www.rolanddga.com/products/3D/modelers/

BR
/T

[miljnor]

How do you assemble and make parts without the use of your arms? do you have someone to help?

If you can pull off manufacturing a cnc mill as a quadriplegic you are my hero. Man you make me realy hate some of the people i have working for me!

your the man!

[Jim DuBois]

Eitanwaks,

If you are only planning to use your machine for cutting something like MDF (frequently used for speaker enclosures in the USA) then IMHO you are significantly over engineered. Like you, I have frequently overbuilt because I used surplus materials available to me at reasonable prices. But, over engineering has it's own set of issues. More mass requires more energy to move it, or it moves slower, for example....as an engineering student you are closer to that than am I.....but moving heaver loads then requires larger motors and drivers....In my experience a gantry approach works very well for CNC routers, plasma cutters, water jets, circut board routers, CNC drills, and the like where side loads are minimal to non existant. Of course part of the exercise is to reduce the mass of the gantry to minimal levels. I have built several gantry machines and several moving table machines. Each has certain advantages and disavantages. Triangulation of stressed components is more efficent than adding more elements to a square or retangular structure, and is usually lighter also. While mass can reduce vibration, vibration is seldom an issue in what is essentially a wood router. Fiberglass, plastics, wood, MDF, and frequently aluminum can be cut in one of these devices. Cutting aluminum will require coolant/lubrication and usually carbide cutting tools....as a general rule all will require an effective chip capture and removal system. I don't find any photos right now of some of my machines, it has been several years since I built my last. If and when I find some I will load them to the site. That may be more helpful than a lot of words and typing, eh?

[eitanwaks]

Piobaire,
if my machine turns out to be anything as good-looking as the one you referred to I will be very happy! You mention vibrations. What is a good way to minimize vibrations? Do you suppose I should change the aluminum excursions for steel ones? They have a higher yield strength therefore will resist deformation better than aluminum.

miljnor,
if you are referring to the building of the machine itself, yes I do have help. I cannot connect all the things together by myself. I do however supervise everything to make sure it is done correctly. I come from a family of engineers and machinists (my great-grandfather, my grandfather, my uncle and father, and now me) so I know little bit about how it should be done. Furthermore, before I was injured I used to work in my grandfather's shop. Regarding work holding, I will probably have someone constrain the workpiece to the table. The beauty about CNC (for me at least) is that I can do 99% of it by myself, that is part modeling in solid works, creating a G. code (CAM software yet to be determined), and operating the machine from a computer.

Jim DuBois,
I notice that I forgot to mention that the speaker cabinets are made of MDF. Nice guest on your part. The front and rear baffle however, is made of aluminum.
The reason I chose to move the table rather than again tree is exactly because of what you mentioned, moving mass requires power.
I agree that triangulation is a better method of dealing with forces and moments however space requirements do not allow for triangulation of the rear end. If you notice, the front side has been dealt with this method.
I suppose I want to cut materials ranging from wood to acrylic to aluminum to steal. This is a reason that I am building it very sturdy. I suppose that something can't be too strong, it can only be not strong enough.

[eitanwaks]

I'm trying to add the SolidWorks file that I made but it is larger than 2 MB. Whenever I try to attach it I get an error because of its size. Is there anyway to attach a file that is larger than 2 MB (the actual file size in zip format is either 4 MB or 11 MB depending on how I save it in SolidWorks)